278 Bayliss and Starling, 



niscus. This latter is seen clearly to consist of three phases: 1^*., a 

 „spike" towards the point of the capillary; 2"''., a more prolonged 

 excursion away from the point; and, 3^'\, a large and prolonged 

 movement towards the point again. Now the connections of the elec- 

 trometer were such that a movement towards the point indicated ne- 

 gativity of the electrode nearest the base of the heart: the curve 

 therefore shows that the hase becomes negative hefore the ax^ex, and 

 that its negativity outlasts that of the apex; apical negativity being 

 indicated by movement away from the point. Our photographs then 

 confirm the first account given by Waller of the variation in man, 

 there being two excursions of the meniscus in the same direction 

 accompanying each beat. 



We have led off and photographed the variations from the va- 

 rious points mentioned by Waller; but found invariably that the elec- 

 trode nearest the base became negative before that nearest the apex. 



Fig. 24 of Plate XVII gives the variations from apex beat and 

 right hand in another person (W. M. B.), and is given to show that 

 here also, movement towards the point (i. e., base negativity) pre- 

 cedes apex negativity; the latter in this case being shewn by the return 

 of the sharp spike to the base line before the commencement of the 

 third phase. 



Fig. 25 gives the variation in the dog under morphia, led off 

 from apex beat and right forepaw. It also is of the same triphasic 

 nature, the base becoming negative before the apex, and remaining 

 negative longer. 



That the normal variation in the intact animal should be triphasic 

 was rather surprising to us, as we had learned to look upon this 

 form of curve in the exposed heart of the dog, as due to local cool- 

 ing of the base. The question suggests itself whether this form of 

 the electrical variation may be due to the proximity of the apex 

 to the warm liver. Against this view is the fact that we found that 

 placing the apex electrode in the dog between the liver and diaphragm, 

 and hence separating the heart from contact with the liver, did not 

 affect the ti-iphasic nature of the variation. Probably therefore, we 



